Modelling a forest lepidopteran: phenological plasticity determines voltinism which influences population dynamics

Abstract

Mnesampela privata is an Australian geometrid moth that is considered to have resource-driven outbreaks. An autumnal oviposition/larval development cycle is considered the “norm” in this species, but spring/summer activity has also been observed. This apparent plasticity of phenology and probable concomitant changes in voltinism have not been considered as possible causes of moth outbreaks. We developed GumMoth, a phenological model for M. privata that uses temperature to predict development times of immatures. Photoperiod determines whether individuals undergo direct development or pupal diapause. We used known catch dates of moths (in the same moth-active season or 365-day period) to simulate population phenologies on the Australian mainland (27°28′S to 38°14′S) and in Tasmania (40°54′S to 42°57′S). GumMoth successfully simulated phenological patterns that agreed with published records and demonstrated for the first time that multivoltinism is possible in M. privata. In seven paired simulations using dates for first and last moth catches, the earliest moth activity resulted in the pupae of four out of 11 generations undergoing diapause, whereas the latest moth activity resulted in pupae of five out of eight generations undergoing diapause (diapause individuals emerged the following moth-active season). Almost two-thirds of offspring of spring/summer active moths reached adulthood within that same moth-active season (i.e. no delay in development), whereas those of autumn active moths always reached adulthood the following moth-active season (i.e. development was delayed). Records from foresters and in the scientific literature confirmed that the more common autumn phenology is supplemented by spring/summer activity in many localities. At half of these localities, populations that required insecticide control or caused substantial defoliation also developed. Because plantation eucalypts are suitable M. privata hosts for only four to five tree-growth seasons after planting, local moth populations must increase quickly if an outbreak is to occur. We suggest that large populations are more likely when plantations are colonised in the first moth-active season after planting and population growth commences in the spring/summer of each subsequent year. Foresters can use catch dates and regional weather data in GumMoth to estimate the number and phenology of generations of M. privata. This information can be used to assess risk posed by regional populations and thereby adjust future surveillance intensity.